Molecular Immunology,Vol. 29, No. 2, pp. 151-159, 1992 Printed in Great Britain.

0161-5890/92 $5.00 + 0.00 Pergamon Press plc

ORGANIZATION OF IMMUNOGLOBULIN HEAVY CHAIN CONSTANT AND JOINING REGION GENES IN THE CHANNEL CATFISH* SEYEDH. GHAFFARI and CRAIG J. LOBBY

Department

of Microbiology,

University of Mississippi Medical Center, Jackson, MI 39216-4505, U.S.A.

(First received 25 February 1991; accepted in revised form 20

June 1991)

Abstract-A channel catfish genomic lambda library was screened with CH and JH probes which were derived from our earlier sequence analyses on different full-length heavy chain cDNA clones. One clone, designated C7, contained a genomic insert of about 18 kb and hybridized with specific probes for each of the four domains of the known C region gene as well as with different oligonucleotides specific for JH gene segments. Southern blot hybridization analysis identified a cluster of JH gene segments which are closely linked to the CH gene. Sequence analysis of the CH-proximal JH element, located about 1.9 kb upstream from the CHl domain, showed that this element contains Y-recombination signals typical of JH elements defined in higher vertebrates, i.e. a nonamer, a 24 bp spacer, and a heptamer. The coding region of this JH element was identical to that contained in the variable region sequence of a cDNA clone previously reported. Sequence analysis of the catfish JH-CH intron suggests that several sequences are present which appear similar to important transcriptional regulatory elements found within JH-CH introns of higher vertebrates. These features include sequences similar to higher vertebrate enhancer elements and regulatory octamers. An additional feature reminiscent of some higher vertebrate heavy chain switch regions is a repetitive sequence area composed of tandemly repeated simple sequences. Lastly, several restriction length polymorphisms were identified and mapped within a 1 kb region located immediately upstream from the JH cluster. This finding suggests that polymorphisms within the IgH locus should be useful in the analyses of channel catfish populations. These combined studies provide further evidence that the genomic organization of heavy chain genes in bony fish shares common organizational features with those known from higher vertebrates.

INTRODUCTION

The mammalian IgH locus contains multiple VH, DH, and JH segments and a single copy of each functional CH gene. For each of the CH genes, exons encode each of the basic structural domains as well as the associated hinge and transmembrane regions. During B cell development, the recombination of the V region gene segments is mediated by specific recombination signal sequences and is accompanied by the processes of somatic mutation, junctional diversity, and N region additions. The functionally rearranged V region is then expressed in association with a C region gene. Through these complex genomic rearrangements, the inherent structural diversity of Ig H chains arises (reviewed by Max, 1989). Phylogenetic studies have shown that although VH, DH, and JH gene segments appear to be conserved in

*This work was supported by a grant from the National Institutes of Health, AI-23052, U.S.A. The nucleotide sequence data reported in this paper has been submitted to the EMBL/GenBank database under the accession number M74041. TAddress correspondence to Dr Craig J. Lobb, Department of Microbiology, University of Mississippi Medical Center, 2500 North State Street, Jackson, MS 39216-4505, U.S.A.

vertebrate phylogeny, there are different genomic organizational patterns which exist to provide antibody diversity. For example, in the chicken, there is only a single JH segment and single functional VH segment (Reynaud et al., 1989). Antibody diversity in this species is achieved by a hyperconversion mechanism wherein the functional VH gene undergoes segmental gene conversion between an extensive VH pseudogene pool. In contrast, studies with the horned shark have shown that VH, DH, JH, and CH gene segments are found in multiple clusters within the genome. Sequence analyses suggest that each of the gene segments in one cluster is highly related to similar segments in other clusters. For example, pairwise comparisons of nine VH segments indicate approximately 86% overall nucleotide sequence similarity (Kokubu et al., 1988). Clearly the genomic arrangement of the Ig heavy chain genes in these lower vertebrates is quite different from that known in higher vertebrates. Our analyses with the channel catfish, however, have shown that the genomic arrangements of H chain genes are distinct from that found in sharks (Ghaffari and Lobb 1989a, 6). Quantitative gene titration experiments have shown that there is a single genomic copy of the catfish CH, gene which is expressed in characterized full length H chain cDNA. Sequence analyses have shown this CH gene is composed of four CH domains and that the CHl and CH4 domains are the most phylogeneti151

S. H.

1.52

GHAFFARI and C. J. LOBB

tally conserved. Genomic blot analyses also indicate that there is likely another CH gene which shares sequence similarity with the characterized gene because probes derived from the CH 1 and CH2 domains cross-hybridize under high stringency conditions. Although it is not known at this time whether this additional gene is functional, earlier studies have shown that antigenically different isotypes of catfish H chains do exist (Lobb and Olson, 1988). Sequence comparisons with higher vertebrate V regions showed that the catfish V region is likely the product of multiple genes with VH, DH, JH elements likely represented. With these findings a central question was whether structural diversity of V region genes co-evolved with the phylogeny of single copy Ig C region genes. In a recent report we showed by sequence relationships and hybridization analyses that an extensive repertoire of VH genes can contribute to antibody diversity in the channel catfish (Ghaffari and Lobb, 1991). Five different groups of VH genes have been identified whose definition is consistent with that of five different VH families. Genomic Southern blots indicate that as many as 100 different germ-line VH genes are likely represented by these VH families. Sequence comparisons of members of the same VH family show that although the framework regions are conserved there is diversity in the CDR regions. For example, in the comparison of three members of the catfish VH2 family, the overall framework similarity at the predicted amino acid level was greater than 92% whereas the combined CDR diversity ranged from 52% to 76%. Additional diversity also appears to be present within the V region likely encoded by JH elements. Comparisons suggest that at least five different genomic JH elements exist which are expressed in members of different VH families. Thus with these analyses showing that there is extensive V region diversity, it was important to understand the genomic arrangement of the catfish V region and C region H chain genes. Earlier genomic restriction mapping analyses had predicted that the four domains of the sequenced catfish CH gene were closely linked upon a genomic fragment less than 6 kb wherein the domain exons were separated by introns (Ghaffari and Lobb, 1989b). In the present report a genomic clone which hybridized with both CH and JH probes was isolated and analyzed. MATERIALS AND METHODS Construction and screening of genomic libraries Genomic DNA was isolated from the erythrocytes of individual channel catfish (Ictalurus p~nctat~~ and purified by methods previously described (Ghaffari and Lobb, 1989b). The purified DNA from one fish was partially digested with MboI and subsequently size fractionated by sucrose gradient centrifugation (Maniatis et al., 1982). DNA fragments of 10-17 kb and 18-24 kb were ligated to BamHI-digested lambdaDASH II arms (Stratogene Cloning Systems, La Jolla, CA) and packaged in vitro. The amplified phage libraries

were screened with both of the following channel catfish heavy chain cDNA inserts: a BstEII-PstI fragment from clone NG13 which encodes the CHl and CH2 domains (Ghaffari and Lobb, 1989b) and a BarnHI-Hind111 fragment from clone HG103 which encodes the CH3, CH4, C-terminal and 3’untranslated region (Ghaffari and Lobb, 1989a). Restr~et~on mapping and sequence analysis

To identify the catfish JH locus, positive genomic clones identified from the above approach were screened with two synthetic oligonucleotides. The first, designated 0521, was a 21 base oligomer (S-GACTACTGGGGAAAAGGAACT-3’) which corresponded to the coding strand of the JH region identified in cDNA clones NG41 and NG64 whereas the second, designated 0523, was a mixed 23 base oligomer which corresponded to the non-coding strand of all catfish JH regions presently defined in cDNA analyses (GTT/GCCT/ CTTT/CCCCCAGTAG~ATCG~AAA; Ghaffari and Lobb, 1991). From 10 positive clones of interest one clone designated clone C7 was chosen for additional study. An ordered restriction map of clone C7 was generated with the aid of a non-radioactive gene mapping kit using alkaline-phosphatase conjugated T3 and T7 oligonucleotide probes (Stratogene Cloning Systems). The restricted DNA from clone C7 was also analyzed by Southern blotting approaches as previously described with the following radiolabeled cDNA probes: CHl, a 335 bp BstEII-SstI fragment from clone NG13; CH2, a 248 bp SstI-EcoRI fragment from clone NG13; CH3, a 271 bp EcoRI fragment from clone HG103; and CH4, a 578 bp EcoRI fragment from clone HG-103 which also encoded the C-terminus and 3’-untranslated region (Ghaffari and Lobb, 1989a, 6). The cDNA fragments were radiolabeled by random priming and the oligomers were end-labeled with T4-polynucleotide kinase as described previously (Ghaffari and Lobb, 1989a). A 3.7 kb XbaI-Hind111 region composed of a 0.8 XbaI fragment and a 2.9 kb XbaI-Hind111 fragment from clone C7 was subcloned into MI3mpl8 and ml3mpl9. The XbaI fragment was sequenced in both directions by the chain te~ination approach using Ml 3 sequencing primers and Sequenase (version 2.0, U.S. Biochemical, Cleveland, OH). Overlapping unidirectional deletion subclones of one strand of the XbaI-HindIII fragment were obtained using exonuclease III (Erase-A-Base, Promega, Madison, WI). The other strand was sequenced using different oligonucleotide primers beginning with a sequencing primer derived from the CHl domain as previously described (Ghaffari and Lobb, 1991). RESULTS AND DISCUSSION ~dent~~~atio~ of genomic clones containing the charnel cat$sh CH gene A genomic lambda library was constructed from the DNA obtained from the erythrocytes of an individual channel catfish. The library was screened with radiolabeled probes representing each of the four catfish CH

Organization JH

CHl

EPHSXhBX

EPHSXhBX

153

of catfish heavy chain genes CH2 E PHSXhBX

CH3 E PHSXhBX

CH4 EPHBXhBX

Fig. 1. Southern blot analysis of channel catfish genomic lambda clone C7. The genomic insert from clone C7 was restricted with seven different enzymes and duplicate blots were respectively hybridized with radiolabeled probes which were specific for JH gene segments (OJ22, a mixed 23-mer) and each of the channel catfish CH domains (CHl, CH2, CH3, CH4). The restriction enzymes used in these

analyses are indicated at the top of each blot (E. EcoRI; P, PstI; H, HindIII; S, SstI; Xh, XhoI; B, BstEII; X, XbaI). domains which were characterized in earlier cDNA sequence analyses (Ghaffari and Lobb, 1989a, b). Several clones were identified, and these clones were screened with two JH-specific oligonucleotides. One clone, designated C7, which contained an insert of about 18 kb was chosen for further analysis because it hybridized with each CH domain probe as well as with both JH probes. To determine if clone C7 contained the genomic insert for the CH gene known from cDNA analyses or the putative second CH gene which was recognized in earlier genomic blot analyses, clone C7 was mapped with eight restriction enzymes, four of which were known to have sites within the cDNA CH domains. Southern blots of the restricted genomic insert were also hybridized with probes specific for each of the CH domains. These results allowed the locations of the four CH domains to be identified and provided information regarding intron sizes (Figs 1 and 2). In cDNA, EcoRI sites are found in the CH2, CH3, and CH4 domains. The EcoRI sites in the CH2 and CH3 domains are separated by 276 bp, whereas the sites in the CH3 and CH4 domains are separated by 81 bp. Southern blot analyses showed that the C7 EcoRI fragment which hybridized with the respective 276 bp EcoRI cDNA fragment was about 700 bp. A C7 300 bp EcoRI fragment was detected when Southern blots were hybridized with a 19 base oligomer which corresponded to an internal sequence within the 81 bp EcoRI cDNA fragment (data not shown). Therefore, the intron between CH2 and CH3 should be about 424 bp whereas the intron which separates CH3 and CH4 should be about 219 bp; a conclusion which is consistent with our earlier genomic restriction analyses (Ghaffari and Lobb, 1989b). When this information is compared with the data of Wilson et al. (1990) who reported the genomic sequence of the CH gene known from cDNA analyses, it is clear that clone C7 contains the same CH gene. Sequence analysis of clone 12c showed the intron between the CH 1

and CH2 domains was 1767 bp, the intron between the CH2 and CH3 domains was 447 bp and the intron between the CH3 and CH4 domains was 234 bp (Wilson et al., 1990). Therefore, both of these clones contain the same length introns which separate the CH domains and the experimentally determined restriction sites defined in clone C7 are mapped to the same positions in the CH locus as those defined in clone 12~. Comparison of the restriction maps of clones C7 and 12c indicates that clone C7 extends an additional 9.2 kb upstream from the 5’-end of clone 12~. Structural analyses and phylogenetic comparisons of the CH domains as well as comparisons of the catfish CH locus with other vertebrate CH loci were reported earlier (Ghaffari and Lobb, 1989a, b; Wilson et al., 1990). Linkage of JH genes segments to the CH locus Our earlier sequence analyses had concluded that there were at least five different JH gene segments which had been expressed in different cDNA clones. Based upon sequence comparisons with known vertebrate JH gene segments, catfish JH segments likely encoded part of the CDR3 as well as the entire FR4 region in the expressed H chains. In addition, it was possible that two other JH segments might exist which differed from one of these five JH segments by a single nucleotide (Ghaffari and Lobb, 1991). To locate the JH locus, two oligonucleotides were synthesized. The first, designated 0523, was a mixed probe which should hybridize to all likely JH gene segments that had been defined in cDNA analyses, whereas the second, designated 0521, should hybridize to the JH gene segment which was expressed in cDNA clones NG41 and NG64 (Ghaffari and Lobb, 1991). Southern blots of the restricted C7 genomic insert were hybridized with 0523, and these results allowed the location of the JH locus to be determined. The 0523 probe hybridized with each of the following C7 genomic restriction fragments: EcoRI 7.8, PstI 15.8, Hind111 9.0

I

I

I

Xbal

BstEll

Xhol

Sst

Hind HI

Pst

EcoR

T7

I

sa

1 kb

I

E

I

I

8

111

E*

I

.

E

JH Cluster

III1

BXHH I

*

JH

I

X

CHt

I

B :.:.:.:. ::::

H

H

E

CH2

S

CH3

XXhESEP

CH4

H

Fig. 2. Restriction enzyme cleavage map and location of JH and CH genes in the genomic channel catfish DNA contained in recombinant lambda clone C7. The restriction sites are designated by the same abbreviations used in Fig. 1 with the addition of SalI (Sa). The locations of the four CH domains (stippled boxes) are indicated and were determined by Southern blot analyses using domain specific probes. The location of the restriction fragments which hybridized with JH specific oligonucleotides (0521 and 0523) is shown by a horizontal arrow; this region is designated the JH cluster. The location of the sequenced JH gene segment in the 0.8 XbaI fragment is indicated by a solid box. The locations of three additional restriction sites are shown by vertical arrows. These additional sites represent restriction site polymorphisms which were defined in individual channel catfish (see Fig. 5). Flanking regions of the lambda DASH II vector (striped) contain the T3 and T7 promoters.

I

I

H

P* E*

T3

E SaSa

Organization of catfish heavy chain genes and 3.4, SstI 14.3, XhoI 15.1, BstEII 3.5 and 2.8, and XbaI 9.9 and 0.8 (Fig. 1). Thus based upon the restriction map of clone C7, all JH hybridizing fragments can be mapped to a 3.9 kb region which would be contained on a EcoRI-XbaI fragment. This region is located about 1.5 kb upstream from the CHl domain (Fig. 2). Several lines of evidence show that multiple JH gene segments are located in this 3.9 kb region. When 0521 was hybridized to Southern blots containing genomic restriction fragments identical to that depicted in Fig. 1, the Hind111 3.4, XbaI 0.8, and BstEII 2.8 fragments did not hybridize whereas all other fragments which hybridized with 0523 also hybridized with OJ21 (data not

401

155

shown). This finding indicates that the JH segment which hybridizes with 0521 is located upstream from the OJ23-positive XbaI 0.8 kb fragment. Secondly, adjacent restriction fragments which overlap this 3.9 kb region hybridized with the 0523 probe e.g. BstEII 3.5 and 2.8; XbaI 9.9 and 0.8; and Hind111 9.0 and 3.4. Densitometric analyses of these adjacent fragments showed that the intensity of these fragments was not equal. The density ratio of the XbaI 9.9 to the XbaI 0.8 fragment was 5.8 whereas the density ratio of the BstEII 3.5 to the BstEII 2.8 fragment was 5.4. Because the XbaI 0.8 fragment contains only one JH gene segment (see below) it is likely that there is a total of 6 or 7 JH segments which are

GTAAGTAATTCAGCGCTTTTACCGTAAACTTTATT

501

TCTCTGCTCATAAAGTGATATATAATGGTCTCAAATGTAA~A*ATTTAT~GGA~GTTAA*A*GT*TTTA~AT**TGTT~ATAA*A*T~A*

601

TTTTTCATCGTTATCAGATTG~G*GACCATTATATGTATAGTAG~~T*AGTG*A*T*TTATTATT**TT*TG~ATTTA*TGTTTTG~T~TTG~TT*AA*~ ,..............

701

*TATTTGTGTAGGCTAATGGTGT*AGTTGTGTGTTT~G~TTAT~~~~*TTTAGAA*~TA*TTT***TGATGT*TT~AG~T~~TG~TTTG~G~~T*T

801

CCGCTTTATACGAGTCAGTTGGTTAATCTATGTAAATCT~A~~G*TAGA~~~TGTG*~GGATT*TTT~TGTGAA*G****~~~~

901

GACAAATTATTTGATTGGTTTAGTTTATTCTAATATGGTAA~T~~T*AT*A~A~A~~GTTTTA*GTTATGA~TA~**~A*TT~A~*~TTAT*~AT~TA~T

1001

ACTAATTATAATAATAACAATAATAATAAGAAGAAAAAGAAGTAGAAATAATATTATAGCCTATAACATTAAATAATATATAGGCTAGTGTATTTAC

1101

CCTTTTTTATATGCTTACTATGAGTTGCTTGGTTAAGTA~~TAAGTAAT~TAAAATATTTTTA*AAATAATAA~AATT~TAAAAAAAAAAAAAAAATTA

1201

AAAAGCACCGTTGGATTCAATATACTGTAGTGATGAOTT~GTATT~TG~T~T~TT~~GTAGTATAATAATAATAATAATAATGATGATGATGATGA~GAT

1301

GATGATGATGATTATTATTATTATTATTATTATTATTATTATTATAAATTGCTCGATTTCGATGTAAAATAAACAATATTATATTTTAACCC~CAGTACA

1601

TTTTTCCTTCCACAACATCAACGAAATGTAACGAAACTTCTATGTA

1701

TGTATGTATGTATGTATGTATGTATGTATGTGTGTATGTATAGTATGTATGTATGTATGTATGTATAGTATGTATGTATGTATGTATCTATGTATGTATC

1801

TATAGTATGTATGTATGTATGTATGTATGTATGTATGTATGTATGTATAGTATGTATGTATGTATGTATGTAT~TATGTA~GTATAGTATGTATGTATGT

1901

ATGTATGTA~TGCATGTTCGCTATGTATTTATTTGTATGTTTAT~AAGAAATAGAATTGATTG~TG~TGTTATAGGCCTATTTATGGATGAATAAT ..............

2001

AATTATTAATCATTTGTATTAATGGATTGTAGATTATTAATAATTTAATGTATTATATATTACGTACTTAAACGGTGAGTAAAATGGAGTACAGTGAAAA ..............

2101

TCTAGGTCTGTTATCTGATTAGAAAGAAGAATAAAATCCATAGAGAAAAACAATGTTGATTGCAAAAAAATAAAATAAAAATCTT

2201

GCATTTTTGGCTTCCAGGCGCAGATTAAATGTTGGTTTGTGCTCTGGCTATCTCTCTCTCTCTCTCTCTCTCTCTCTCTCTAACTATCTATATCTACAGC

2301

CHl TGTGCAAAGCGCCCCG ValGlnSerAlaPro

.l-

Fig. 3. Nucleotide sequence of the channel catfish CHl-proximal JH gene segment and JH-CHl intron. The sequence begins at the XbaI site preceding the JH element (see Fig. 2) and extends through the CHl domain (only the first five codons of the CHl domain are shown). Restriction sites within this region that are indicated in Fig. 2 are found at the following nucleotide positions: XbaI, 1 and 838; HindIII, 304 and 397. The likely JH heptamer and nonamer recombination signal sequences are boxed and were derived by sequence comparisons with similar elements identified in higher vertebrate JH gene segments. The amino acid sequence of the JH coding region (also boxed) was identified by comparisons to JH-encoded regions in different channel catfish cDNA clones (Ghaffari and Lobb, 1991). Nucleotide sequences sharing similarity to transcriptional regulatory elements located within the JH-CH intron of higher vertebrates are indicated by solid (enhancer elements) and dotted underlining (regulatory octamer). The sequence area characterized by the repeated tetranucleotide TGTA is indicated by arrows and extends from nucleotide position 1681 through position 1914. The demarcation of the coding region of the CHl domain was determined by comparisons with previous analyses on full length catfish heavy chain cDNA (Ghaffari and Lobb, 1989a, 1989b).

156

S. H. GHAFFARIand C. J. L~BB CDR3

CDNA (NG77)

340

I

I

PR4 I

T’

N D G K T G;C A:G Ci!TAAC T!G G;T T:T GAT T:C T!G GGA AAA G:C ACA GtT G;C ACC G%

cm '380 I

A&

s AGC GtT G;G $A

A&

G&

C:G

. . . . . . . . . . . . . ..ATCTAACTATCTATATCTACCAgCT G;G CL A:, G;, C:G Fig. 4. Comparison of the CDR3/FR4/CHl region in channel catfish cDNA clone NG77 with JH and CHl genomic sequences. The full-length cDNA sequence of clone NG77 was reported earlier (Ghaffari and Lobb, 1991); the genomic sequence of the CH-proximal JH element is depicted beginning with the heptamer sequence. The respective RNA splice sites in the JH and CHl genomic sequences are indicated by brackets.

located in this 3.9 kb region; an estimate which is in close agreement with the predicted number of different JH segments defined by cDNA sequence analyses. Sequence

FR4 360

u P D Y U G K G T V T V T S .CACTGTGT AiC TGG GtT TTT GAT TAC TGG GGA AAA GGC ACA G$T GTC ACC GTA ACC AGC G@L4GTMTTCAGCG......

JH

CHl

320

analysis

of the CH proximal

JH element

With these results indicating that the XbaI 0.8 restriction fragment should contain a JH gene segment, a 3.7 kb XbaI-Hind111 fragment which contained the 0.8 kb XbaI fragment and extended downstream to contain the CHl domain was chosen for sequence analysis. The nucleotide sequence of this region extending into the CHl domain is shown in Fig. 3. Sequence analysis showed that this region contained one JH gene segment. This JH segment has a 5’-recombination signal sequence which is typical of that associated with other vertebrate JH segments: a nonamer, a 24 bp spacer, and a heptamer. The comparison of the 5’-recombination signal sequence upstream of JH segments defined in human (Ravetch et al., 1981), mouse (Sakano et al., 1980), rabbit (Becker et al., 1989), Xenopus (Schwager et al., 1988), ladyfish (Amemiya and Litman, 1990), and horned shark (Kokubu et af., 1988) enable several conclusions to be made. Comparisons of the heptamer and nonamer sequence indicated that this heptamer is generally more conserved than the nonamer. Although none of these other species have nonamer sequences identical to this catfish JH nonamer, the same JH heptamer has been identified in some JH segments from Xenopus and the horned shark. The length of the spacer regions between the heptamer and nonamer sequences in these various species ranges from 21 to 25 bp, whereas the likely spacer region in this catfish JH segment is 24 bp. This 24 bp spacer shares no obvious sequence similarity with the spacer regions identified in these other vertebrate JH segments. The presence of a 24 bp spacer predicts that DH elements which undergo recombination with this JH element would likely contain 12 bp spacers between their heptamer and nonamer elements so as to maintain the 12/23 bp spacer rule (Early et al., 1980; Sakano et al., 1980; Sakano et al., 1981). Of additional interest is that a 17 bp sequence (Fig. 3, nucleotides 366382) which includes the JH nonamer is repeated immediately upstream. This reiterated sequence (nucleotides 341-357) contains an additional copy of the JH nonamer. The functional significance of two identical nonamers separated by 16 bp is not known.

The coding sequence in this JH gene segment was compared to previously sequenced cDNA clones to determine if the coding region had likely been expressed. In the 11 catfish heavy chain cDNA clones whose sequence was reported earlier (Ghaffari and Lobb, 19896, 1991), one clone, designated NG77, contained the same coding sequence as that defined in the JH element. The CDR3/FR4/CHl sequence in this cDNA clone was compared with the genomic sequence of the JH segment and the sequence of the upstream flanking region of the CHl exon. This comparison shows several features (Fig. 4). This JH element likely encodes all of the FR4 region as well as six of the 14 residues of the CDR3 region. The other eight residues in the CDR3 are likely contributed by VH and DH elements and may include N region contributions. No somatic mutation likely occurred within the JH encoded region of this cDNA clone; the expressed nucleotide sequence was identical to that determined in the genomic sequence of the JH gene segment. The coding sequence of the JH segment terminates immediately prior to the RNA donor splice site G/GTA, whereas the RNA acceptor splice site CAGjC is found immediately upstream from the coding sequence of the CHl exon. Identification of the catfish CH-proximal JH segment also allows additional comparisons to be made. The JH locus in the human and mouse is respectively located about 6 and 7 kb upstream from the Cy gene (Sakano et al., 1980; Wood and Tonegawa, 1983; Ravetch et al., 1981). There are four functional and one pseudo-JH segment in the mouse, whereas three pseudo-JH elements are interspersed among six functional human JH segments. In Xenopus the JH locus, as presently defined, is located about 8 kb upstream from the Cp gene (Schwager et al., 1988). Six functional JH segments have been identified although cDNA analyses suggest that other JH elements exist (Hsu et al., 1989; Haire et al., 1990). Amemiya and Litman (1990) sequenced a JH element from another teleost, the ladyfish, and showed that this element was separated from the CHl domain by a maximum distance of approximately 3.6 kb. By hybridization analyses this JH element was the only element found within an 11 kb upstream region from the CH gene. Because other non-overlapping lambda clones were identified which hybridized with a JH probe, these investigators suggested that JH elements might be

Organization of catfish heavy chain genes separated by considerable lengths. Based upon the above analyses with the catfish, however, multiple JH segments are closely linked to the characterized CH locus. The CH-proximal JH element is located about 1.9 kb upstream from the CHl domain, a distance much shorter than that defined in each of these other vertebrates. Clearly additional studies to sequence the remaining JH elements identified in the JH cluster will further our knowledge on the organizational patterns of JH gene segments. Sequence analysis of the catjish JH-CH intron The transcription of higher vertebrate Ig genes is dependent upon tissue specific regulatory sequences which are represented by promoter and enhancer elements (reviewed by Sen and Baltimore, 1989). In higher vertebrates a conserved octamer and enhancer elements reside in the introns between JH and CH genes. These sequences are recognized by their ability to control transcription of immunoglobulin genes in transfection assays (Banerji et al., 1983; Gillies et al., 1983). An enhancer core consensus sequence, CAGGTGG, however, has been derived by the analysis of mammalian enhancer elements (Church et al., 1985; Sen and Baltimore, 1989). In the catfish JH-CH intron, three sequences similar to the enhancer core consensus sequence were identified, and these are indicated in Fig. 3 (nucleotide positions 816822, 11261132 and 1208-1214, respectively). The first two sequences contain a core sequence similar to known enhancer elements, and both are represented by the same 11 bp sequence (GTCAGTTGGTT). The third sequence, although more divergent, has a similar core sequence (CCGTTGG). Other studies have shown that the regulatory octamer, which is also found within the intron between JH and CH genes, is also represented by a conserved sequence. This sequence, originally defined as the octanucleotide ATTTGCAT (Parslow et al., 1984) has been shown to consist of nine conserved bases TNATTTGCAT where N represents any base (Wirth et al., 1987); the term octamer, however, has been retained for historical consistency (Sen and Baltimore, 1989). In the catfish JH-CH intron there are three sequences similar to the mammalian octamer (Fig. 3, nucleotide positions 665-675, 1933-1942 and 2010-2019, respectively). This first sequence TAATTATGCAT is identical to the consensus sequence except that it contains a one base insertion (A is inserted in the middle of the trinucleotide sequence TTT of the consensus octamer). The other two sequences (TTATTTGTAT and TCATTTGTAT, respectively) are identical to the consensus octamer except at one position. Inspection of sequence areas surrounding these potential catfish enhancer and octamer sequence, however, shows no obvious sequence relationships with JH-CH intron sequences from other vertebrates which contain these elements. Clearly, functional assays will need to be conducted prior to concluding that these elements have regulatory transcriptional function. An additional feature found within the catfish JH-CH intron is a 30 bp TC rich area found immediately

157

preceding the CHl domain (Fig. 3 nucleotide positions 2251-2281). In other systems such regions have been shown to bind proteins, termed GAGA factors, which are suggested to be transcriptional antirepressors (reviewed by Kerrigan et al., 1991; Croston et al., 1991). The significance of this TC rich region in the catfish JH-CH intron, however, is not known. It should also be noted that Wilson et al. (1990) reported about 100 nucleotides upstream of the CHl domain during their genomic sequence analysis of the CH gene. Comparison shows that this TC rich area in their clone 12c was comprised of approximately 50 nucleotides rather than the 30 nucleotides which were present in clone C7. Although comparisons show that upstream and downstream sequences surrounding this TC rich area are identical in both clones, it is not known whether this difference reflects individual variability between channel catfish or stability of this region during cloning. A last feature is that in the catfish JH-CH intron there is a relatively long repetitive sequence area characterized by the repeated tetranucleotide TGTA (Fig. 3, nucleotide positions 1681-1914). Various lengths of the repeated tetranucleotide sequence appear to be punctuated by a pentanucleotide bearing a one base insertion (TAGTA). This general sequence motif is repeated 6 times and hence could be characterized as [(TGTA), TAGTA], where n is 15, 5, 8, 10, 8 and 5, respectively. In higher vertebrates some repetitive sequence regions located within the JH-CH intron are known to be associated with heavy chain recombination switch (S) regions. These regions are also often composed of simple sequences which are tandemly repeated (reviewed by Honjo et al., 1989). Inspection of characterized S regions in higher vertebrates, however, shows no apparent sequence similarity with this repeated sequence area in the catfish JH-CH intron. Although it is tempting to speculate that heavy chain switch recombination mechanisms might exist in this lower vertebrate, clearly additional analyses including characterization of the putative second CH gene will be required. Restriction length polymorphisms in the catfish IgH locus The genomic DNA from 9 individual catfish was restricted with either PstI or EcoRI and Southern blots were hybridized with radiolabeled probes representing each of the four CH domains. These analyses showed that restriction length polymorphisms exist in the catfish IgH locus (Fig. 5). With PstI restricted DNA, either three or four hybridizing bands were detected. Neither the PstI 5.6 band (which contains about 300 bp of the CH4 domain, the C-terminus of the secreted H chain, and downstream flanking region), nor the PstI 3.1 band (which was previously shown to cross-hybridize with probes derived from the CHl domain and suggested to represent a second CH gene, Ghaffari and Lobb, 1989b), exhibited polymorphic variation. However, the PstI fragment which contains the CHl, CH2, CH3, and about 100 bp of the CH4 domain was polymorphic and two size variants were observed: a 10 kb fragment and a fragment designated 16 kb (a minimal size estimate). In

S. H. GHAFFARI and C. J. LOBB

158

Pst I 123456789

EcoRl 123456789

16.0,a.4 -8.0 -7.8

5.6 - 5.0

-

1.9

- 0.7 Fig. 5. Southern hybridized with correspond to polymorphisms

blots of EcoRI or PstI restricted genomic DNA channel catfish probes for each of the four CH the DNA from the individual fish. The sizes (PstI 16.0 and 10.0; EcoRI 8.4 and 8.0 and lambda DNA markers.

some fish only the 10 kb fragment was detected (fish 4 and 7) whereas in other fish only the 16 kb fragment was detected (fish 1, 6, 8, and 9). In other fish, however, both the 10 and 16 kb fragments were observed (fish 2, 3, and 5). Densitometric analysis showed that the density of the 10 and 16 kb bands in the fish which exhibited both bands was essentially equal and this combined density was essentially equal to the density of either the 10 kb or 16 kb fragments identified in those fish which did not exhibit both bands. These analyses are consistent with the conclusion that the PstI 10 kb and PstI 16 kb fragments are found on different alleles. EcoRI digests of the genomic DNA from these same fish also showed polymorphisms. In all fish four or five bands were observed which included monomorphic fragments of 0.7, 1.9 and 5.0 kb. The 0.7 kb fragment contains the CH3 domain; the 1.9 kb fragment contains the CH4, C-terminus, and the downstream flanking region; and the 5.0 kb fragment likely represents the uncharacterized second CH gene. In contrast the fragment which contains the upstream flanking region and the CHl and CH2 domains was represented by three different polymorphic fragments of 7.8, 8.0, and 8.4 kb. In some fish only the EcoRI 7.8 was detected (fish 1, 2, 6, 8, 9) whereas in other fish two EcoRI fragments were observed. Fish 3 and 7 had both the EcoRI 8.0 and 8.4 fragments, whereas fish 4 and 5 had both the EcoRI 7.8 and EcoRI 8.0 fragments. Densitometric analyses supported the conclusion that these fragments represent three different alleles.

from nine individual channel catfish domains. The designated lanes (l-9) of the restriction fragment length 7.8) were estimated from restricted

These combined analyses suggest that there are minimally five alleles which are unequally represented in this small catfish population. The allele with both the PstI 16.0 and EcoRI 7.8 polymorphisms is most abundant and is represented in fish 1, 2, 6, 8, and 9; each of these fish (except fish 2) appears to be homozygous at these allelic positions. When the EcoRI and PstI polymorphisms are mapped on the catfish IgH locus, the EcoRI 7.8, 8.0 and 8.4 and the PstI 10.0 sites are clustered in a region which is less than 1 kb in size and located upstream from the JH locus (see Fig. 2). Localized polymorphisms are often found near highly repetitive sequence areas, and clustered polymorphisms have been identified near CH genes of higher vertebrates (e.g. Migone et al., 1983; Beth-Hansen et al., 1983; Knight et al., 1985; Bottaro et al., 1989). These analyses suggest that these polymorphisms should be useful in the analyses of channel catfish populations similar to those employed with higher vertebrates. Additional sequence analyses are now required to further define this polymorphic area. In conclusion, these studies provide further evidence to support our earlier analyses that the genomic organization of heavy chain genes in bony fish shares common organizational features with those known in higher vertebrates. Bony fishes appear to be the first known representatives to have evolved single copies of C region genes, to differentially express members of different VH families with the same C region gene, and to have multiple JH gene segments which are closely linked to a

Organization

of catfish heavy chain genes

CH gene. These studies should continue to provide important insight into the phylogeny of immunoglobulin structure and function.

159

Kerrigan L. A., Croston G. E., Lira L. M. and Kadonaga J. T. (1991) Sequence-specific transcriptional antirepression of the Drosophila Kriippel gene by the GAGA factor. J. biol. Chem. 266, 574-582.

REFERENCES Amemiya C. T. and Litman G. W. (1990) Complete nucleotide sequence of an immunoglobulin heavy-chain gene and analysis of immunoglobulin gene organization in a primitive teleost species. Proc. natn. Acad. Sci. U.S.A. 87, 811-815. Banerji J., Olson L. and Schaffner W. (1983) A lymphocytespecific cellular enhancer is located downstream of the joining region in immunoglobulin heavy chain genes. Cell 33, 1299740. Beth-Hansen N. T., Linsley P. S. and Cox D. W. (1983) Restriction fragment length polymorphisms associated with immunoglobulin Q gene reveal linkage disequilibrium and genomic organization. Proc. natn. Acad. Sci. U.S.A. 80, 69526956. Becker R. S., Zhai S. K., Currier S. J. and Knight K. L. (1989) Ig VH, DH, and JH germ-line gene segments linked by overlapping cosmid clones of rabbit DNA. J. Zmmun. 142, 1351-1355. Bottaro A., Gallina R., Demarchi M. and Carbonara A. (1989) Genetic analysis of new restriction fragment length polymorphisms (RFLP) in the human IgH constant gene locus. Eur. J. Immun. 19, 2151-2157. Church G. M., Ephrussi A., Gilbert W. and Tonegawa S. (1985) Cell-type-specific contacts to immunoglobulin enhancers in nuclei. Nature 313, 798-801. Croston G. E., Kerrigan L. A., Lira L. M., Marshak D. R. and Kadonaga J. T. (1991) Sequence-specific anti-repression of histone HI-mediated inhibition of basal RNA polymerase III transcription. Science 251, 643-649. Ephrussi A., Church G. M., Tonegawa S. and Gilbert W. (1985) B lineage-specific interactions of an immunoglobulin enhancer with cellular factors in vivo. Science 227, 134140. Ghaffari S. H. and Lobb C. J. (1989~) Cloning and sequence analysis of channel catfish heavy chain cDNA indicate phylogenetic diversity within the IgM immunoglobulin family. J. Immun. 142, 13561365. Ghaffari S. H. and Lobb C. J. (1989b) Nucleotide sequence of channel catfish heavy chain cDNA and genomic blot analyses. Implications for the phylogeny of Ig heavy chains. J. Immun.

143, 2730-2739.

Ghaffari S. H. and Lobb C. J. (1991) Heavy chain variable region gene families evolved early in phylogeny: Immunoglobulin complexity in fish. J. Zmmun. 146, 1037-1046. Gillies S. D., Morrison S. L., Oi V. T. and Tonegawa S. (1983) A tissue specific transcription enhancer element is located in the major intron of a rearranged immunoglobulin heavy chain gene. Cell 33, 717-728. Haire R. N., Amemiya C. T., Suzuki D. and Litman G. W. (1990) Eleven distinct VH gene families and additional patterns of sequence variation suggest a high degree of immunoglobulin complexity in a lower vertebrate, Xenopus laevis. J. exp. Med. 171, 1721-1737. Honjo T., Shimizu A. and Yaoita Y. (1989) Constant-region genes of the immunoglobulin heavy chain and the molecular mechanism of class switching. In Immunoglobulin Genes (Edited by Honjo T., Alt F. W. and Rabbitts T. H.), pp. 123-149. Academic Press, London. Hsu E., Schwager J. and Alt F. W. (1989) Evolution of immunoglobulin genes: VH families in the amphibian Xeno pus. Proc. natn. Acad. Sci. U.S.A. 86, 8010-8014.

Knight K. L., Burnett R. C. and McNicholas J. M. (1985) Organization and polymorphism of rabbit immunoglobulin heavy chain genes. J. Immun. 134, 1245-1250. Kokubu F., Litman R., Shamblott M. J., Hinds K. and Litman G. W. (1988) Diverse organization of immunoglobulin VH gene loci in a primitive vertebrate. EMBO J. 7, 3412-3422.

Lobb C. J. and Olson M. 0. J. (1988) Immunoglobulin heavy chain isotypes in a teleost fish. J. Immun. 141, 12361245. Maniatis T., Fritsch E. F. and Sambrook J. (1982) Molecular Cloning: A Laboratory Manual, p. 284. Cold Spring Harbor Laboratory, Cold Spring Harbor. Max E. E. (1989) Immunoglobulins: Molecular Genetics. In Fundamental Immunology, Second Edition, (Paul E. W.), pp. 235-290. Raven Press, New York. Migone N., Feder J., Cann H., VanWest B., Hwang J., Takahashi N., Honjo T., Piazza A. and Cavalli-Sforza L. L. (1983) Multiple DNA fragment polymorphisms associated with immunoglobulin CIchain switch-like regions in man. Proc. natn. Acad Sci. U.S.A. 80, 467-471.

Parslow T. G., Blair D. L., Murphy W. J. and Granner D. K. (1984) Structure of the 5’ ends of immunoglobulin genes: A novel conserved sequence. Proc. natn. Acad. Sci. U.S.A. 81, 2650-2654.

Ravetch J. V., Siebenlist, U., Korsmeyer, S., Waldmann T. and Leder P. (1981) Structure of the human immunoglobulin p locus: Characterization of embryonic and rearranged J and D genes. CeN 27, 583-591. Reynaud C.-A., Dahan A., Anquez V. and Weill J.-C. (1989) Somatic hyperconversion diversifies the single VH gene of the chicken with a high incidence in the D region. Cell 59, 171-183. Sakano H., Maki R., Kurosawa Y., Roeder Y. and Tonegawa S. (1980) Two types of somatic recombination are necessary for the generation of complete immunoglobulin heavy-chain genes. Nature 286, 676683. Sakano H., Kurosawa Y., Weigert M. and Tonegawa S. (1981) Identification and nucleotide sequence of a diversity DNA segment (D) of immunoglobulin heavy- chain genes. Nature 290, 562-565.

Schwager J., Grossberger D. and DuPasquier L. (1988) Organization and rearrangement of immunoglobulin M genes in the amphibian Xenopus. EMBO J. 7, 2409-2415. Sen R. and Baltimore D. (1989) Factors regulating immunoglobulin-gene transcription. In Immunoglobulin Genes (Edited by Honjo T, Alt F. W. and Rabbitts T. H.), pp. 327-342. Academic Press, London. Wilson M. R., Marcuz A., van Ginkel F., Miller N. W., Clem L. W., Middelton D. and Warr G. W. (1990) The immunoglobulin M heavy chain constant region gene of the channel catfish, Ictalurus punctatus: An unusual mRNA splice pattern produces the membrane form of the molecule. Nucl. Acid. Res. 18, 5227-5233.

Wirth T., Staudt L. and Baltimore D. (1987) An octamer oligonucleotide upstream of a TATA motif is sufficient for lymphoid specific promoter activity. Nature 329, 174-178. Wood C. and Tonegawa S. (1983) Diversity and joining segment of mouse immunoglobulin heavy chain genes are closely linked and in the same orientation: implications for the joining mechanism. Proc. natn. Acad. Sci. U.S.A. 80, 303ck3034.